This invention relates to a method and apparatus for calibrating a radiological image, and in particular an image by an X-ray radiography device.
The images obtained in X-ray radiography are projections on a plane of two-dimensions (the plane of the means for acquiring an image) of three-dimensional volumes. Due to the loss of one dimension, it is not possible to carry out direct measurements on the images acquired if no calibration procedure is provided. FIG. 1 shows schematically an X-ray source S and a projection plane 2. In this figure, the same object 3 will correspond to projections 5 and 6, on the plane 2, of different sizes, depending on its position between the X-ray source S and the projection plane 2. Similarly, two objects 3 and 4 of different sizes may correspond to the same projection 5 on the plane 2. A calibration procedure enables calculating the sizes of elements appearing on the acquired images. Generally, the calibration procedure comprises identifying an object whose dimensions are known, located in the immediate proximity of the region to be measured. Such an object may be a catheter, a sphere or even a grid. Knowing the dimensions of this object, it is possible to determine the dimensions of the elements which appear on the acquired images and which are substantially in the same plane as the object which is acting as a standard. However, this calibration procedure is relatively tedious since each time requires: positioning a standard object, the dimensions of which are known, in a suitable place; acquiring the dimensions of the standard object at the X-ray device; and if necessary, take acquired X-ray images of the object.
This known procedure penalizes the productivity and is, furthermore, harmful to the patient, since the patient will be subject to several X-ray doses for the calibration which will not be directly used in establishing a diagnosis.